Refers to article:
Expression of AR-V7 in Circulating Tumour Cells Does Not Preclude Response to Next Generation Androgen Deprivation Therapy in Patients with Castration Resistant Prostate Cancer
Accepted 14 July 2016
January 2017 (Vol. 71, Issue 1, pages 1 - 3)
A critical unmet need in advanced prostate cancer (PCa) management is how to best sequence the available life-prolonging therapies to maximize clinical benefit for the individual patient and, in particular, to identify those men most likely to respond to a next-generation hormonal agent (abiraterone or enzalutamide) versus a taxane agent (docetaxel or cabazitaxel). In 2014, it was first reported that the detection of androgen receptor splice variant 7 (AR-V7) messenger RNA (mRNA) in circulating tumor cells (CTCs) isolated from the blood of patients about to start a new line of therapy for castration-resistant prostate cancer (CRPC) was associated with a lack of response to abiraterone and enzalutamide . This clinical observation was consistent with the known biological functions of AR-V7 and the mechanism of action of the drugs, and it was subsequently confirmed by multiple groups using a range of laboratory assays . No such association was observed with taxane-based therapies, suggesting that this class of agent would be more likely to benefit patients in whom AR-V7 was detected in CTCs  and .
The report by Bernemann et al  in this month's issue of European Urology, in which the authors conclude that “6 of 21 AR-V7–positive patients experienced benefit” from abiraterone or enzalutamide, appears to contradict the growing body of evidence. A critical review of the findings, however, does not support this conclusion and highlights the need for adherence to the evidentiary standards for biomarker validation (analytical and clinical) and the reporting of treatment outcomes. It also underlines the appropriate concern of regulators and third-party payers about the comparability of the results of laboratory-developed tests (LDTs) for predictive biomarkers developed by different groups, and the reason predictive biomarker assays used to select one form of therapy over another are classified as “class III” devices where every step (from the type of tube used in the initial blood collection, to the reporting of the results) must be locked down prior to clinical qualification .
We first consider the AR-V7 assay that the authors state was “the identical technique used by Antonarakis et al in their 2014 publication.” Both assays were LDTs: one was developed and validated at Johns Hopkins University and the second at University Hospital Muenster. Both used the AdnaTest platform (Qiagen, Hanover, Germany) to capture CTCs and extract RNA, followed by reverse-transcription polymerase chain reaction (PCR) to detect AR-V7 mRNA using customized primers  and . A closer inspection, however, revealed that the primers used to detect AR-V7 transcripts differed between the two assays, as did the PCR conditions used in each test. A Basic Local Alignment Search Tool (BLAST) query (www.ncbi.nlm.nih.gov/blast/) of the PCR primers used in the Bernemann et al assay showed significant hybridization with other genes, raising questions about specificity. Another potentially troublesome feature is that their forward primer straddles the AR-V7 splice junction, that is, the junction between canonical exon 3 (E3) and cryptic exon 3 (CE3) of the AR gene, with 10 base pairs of the forward primer sequence being complementary with the CE3 sequence. Because the reverse primer is also in CE3, this design runs the risk of detecting unprocessed pre-mRNA, or even DNA (if RNA extraction is not clean), unless it can be definitively ruled out experimentally. AR-V7 pre-mRNA may be present (even when AR-V7 mature RNA is absent) in PCa cells as shown recently by RNA in situ hybridization . Therefore, the risk of false-positive results in the current study due to pre-mRNA or DNA detection (rather than true AR-V7 mRNA detection) remains a possibility.
As for the clinical outcomes, “substantial benefit” was defined by the authors as “either stable disease (<50% prostate-specific antigen [PSA] decrease to <25% increase from PSA nadir) or response (PSA decrease >50%).” Not applied were the reporting standards outlined in the Prostate Cancer Clinical Trials Working Group 2 (PCWG2)  and updated in PCWG3 . It is well recognized that transient declines in PSA levels can occur following treatment with AR-directed therapies that are of no clinical significance, and standard reporting requires confirmation of the PSA decline at a minimum of 8 wk (and preferably 12 wk) after treatment is started. The patient who unfortunately succumbed 26 d after initiating therapy is also simply unevaluable for PSA response. No data are included with respect to radiographic changes or radiologic progression-free survival (rPFS), an end point developed in part through PCWG2 and subsequently included in registration trials. In either case, a 3-mo treatment duration would not be considered “beneficial” in our view. It is also difficult to interpret the results presented by Bernemann et al accurately in an unbiased manner without knowledge of the corresponding clinical outcomes of men with AR-V7–negative status as determined using their assay. To this end, data on AR-V7–negative patients treated with novel hormonal therapy were not reported. Nor was it stated how many total patients with CRPC were screened (ie, the denominator) to identify 21 AR-V7–positive cases.
An alternative way to assess AR-V7 status is at the protein level. In a recent study, it was demonstrated that the presence of nuclear-localized AR-V7 protein as detected by an immunofluorescence assay using the Epic Sciences CTC platform (San Diego, CA, USA) was associated with a lack of PSA declines, shorter rPFS, and shorter survival (the latter in multivariable analysis adjusting for known clinical prognostic factors) in men receiving abiraterone and enzalutamide . Even the presence of one AR-V7–positive cell appeared to predict lack of response to these agents. Interestingly, most CTCs in AR-V7–positive blood samples lacked the AR-V7 protein, which might explain the transient PSA declines observed in the present report (recognizing that these declines do not represent a clinical benefit). A treatment-specific interaction between the presence of AR-V7 and response to taxanes was also demonstrated , confirming an earlier study . In addition, the relative reduction in risk of death was 76% (hazard ratio: 0.24; 95% confidence interval, 0.10–0.57), making taxanes a viable, albeit noncurative, treatment option for AR-V7–positive patients. In contrast, the cancer-specific outcomes with these two drug classes appeared comparable in patients with AR-V7–negative CTCs. Taken together, these two studies  and  suggest that treatment with a taxane is more likely to benefit a patient in whom AR-V7 is detected in CTCs at the time that a treatment decision is needed. Both hypotheses can only be answered with dedicated prospective trials focused on the biomarker question.
The authors conclude that “AR-V7 expression alone cannot predict sensitivity to next-generation hormonal therapy,” and we agree. However, its presence in CTCs is highly specific for lack of response to AR-directed therapy, whereas its absence in CTCs does not uniformly predict outcome. Urgently needed are locked-down, analytically validated, and certified assays for AR-V7 (mRNA or protein) that are prospectively validated in trials specifically designed to address the biomarker question that shows clinical utility, namely, that the outcome of the patient is ultimately improved through knowledge of the test result (biomarker-informed approach) relative to an uninformed (biomarker-agnostic) approach. Showing prospectively that patients with AR-V7 in CTCs may benefit more from taxanes relative to novel hormonal agents is a clear example of clinical utility that (if demonstrated) could be of immediate benefit to patients.
Conflicts of interest
Emmanuel S. Antonarakis has served as a paid consultant/adviser for Janssen, Astellas, Essa, and Medivation; has received research funding to his institution from Janssen, Johnson & Johnson, Medivation, and Tokai; and is a coinventor of a biomarker technology that has been licensed to Tokai. Howard I. Scher has served as a compensated consultant/advisor for Astellas, BIND Pharmaceuticals, Blue Earth Diagnostics, Clovis Oncology, Elseiver's Practice Update Website, Ferring Pharmaceuticals, Med IQ, Merck, Roche, Sanofi and WCG Oncology; is an uncompensated consultant for Medivation; serves on the Board of Directors for Asterias Biotherapeutics; and has received research funding to his institution from Illumnia, Innocrin Pharma, Janssen and Medivation.
-  E.S. Antonarakis, C. Lu, H. Wang, et al. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med. 2014;371:1028-1038 Crossref
-  E.S. Antonarakis, A.J. Armstrong, S.M. Dehm, J. Luo. Androgen receptor variant–driven prostate cancer: clinical implications and therapeutic targeting. Prostate Cancer Prostatic Dis. 2016;19:231-241
-  E.S. Antonarakis, C. Lu, B. Luber, et al. Androgen receptor splice variant 7 and efficacy of taxane chemotherapy in patients with metastatic castration-resistant prostate cancer. JAMA Oncol. 2015;1:582-591 Crossref
-  Scher HI, Lu D, Schreiber NA, et al. Association of AR-V7 on circulating tumor cells as a treatment-specific biomarker with outcomes and survival in castration-resistant prostate cancer. JAMA Oncol. In press. http://doi.org/10.1001/jamaoncol.2016.1828.
-  C. Bernemann, T.J. Schnoeller, M. Luedeke, et al. Expression of AR-V7 in circulating tumour cells does not preclude response to next generation androgen-deprivation therapy in patients with castration-resistant prostate cancer. Eur Urol. 2017;71:1-3
-  G.J. Kelloff, C.C. Sigman, H.I. Scher. Biomarker development in the context of urologic cancers. Urol Oncol. 2015;33:295-301 Crossref
-  Steinestel J, Luedeke M, Arndt A, et al. Detecting predictive androgen receptor modifications in circulating prostate cancer cells. Oncotarget. In press. http://doi.org/10.18632/oncotarget.3925.
-  Guedes LB, Morais CL, Almutairi F, et al. Analytic validation of RNA in situ hybridization (RISH) for AR and AR-V7 expression in human prostate cancer. Clin Cancer Res. In press. http://doi.org/10.1158/1078-0432.CCR-16-0205.
-  Scher HI1, S. Halabi, I. Tannock, et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol. 2008;26:1148-1159
-  H.I. Scher, M.J. Morris, W.M. Stadler, et al. Trial design and objectives for castration-resistant prostate cancer: updated recommendations from the Prostate Cancer Clinical Trials Working Group 3. J Clin Oncol. 2016;34:1402-1418
a Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Johns Hopkins University, Baltimore, MD, USA
b Memorial Sloan Kettering Cancer Center, Sidney Kimmel Center for Prostate and Urologic Cancers, and Weill Cornell Medical College, New York, NY, USA
Corresponding author. Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, 1650 Orleans Street, CRB1–1M45, Baltimore, MD 21287, USA. Tel. +1 443 287 0553; Fax: +1 410 614 8397.
© 2016 European Association of Urology, Published by Elsevier B.V.